A Cooperative Scheduling Control Method for Weeders in Field Environment

Abstract

The present invention relates to the field of automatic control, and particularly to a collaborative scheduling control method for a weeding machine in a large field environment, including using a drone equipped with a camera, a multispectral sensor, and different sensor devices to collect information related to the large field and weeding and preprocess the data, establishing a neural network model for evaluating the priority of weeding operations, and outputting the evaluation result of the priority of weeding operations; obtaining the confidence level of the evaluation information of the large field weeding operation according to the output evaluation information of the priority of weeding operations, setting a credibility threshold for the evaluation information of weeding operations at the same time, and calculating the real-time weeding operation speed by combining the confidence level of the evaluation information of the large field weeding operation with the power consumption; calculating the determination index of the large field weeding efficiency by obtaining the large field weeding operation speed and the performance of the weeding equipment, and if the obtained determination index of the large field weeding efficiency exceeds the range of the weeding benchmark efficiency threshold, starting a preparatory work countdown.

Description

Technical Field

[0001] The present invention relates to the field of automatic control, and particularly to a collaborative scheduling control method for a weeding machine in a large field environment. Background Art

[0002] In modern agricultural production, the planting area of large fields is vast. The growth of weeds will compete with crops for nutrients, water, and sunlight, seriously affecting the yield and quality of crops. Weeding operations are an important part of large field management. The traditional manual weeding method is inefficient, labor-intensive, and costly; while some existing weeding machines mostly operate independently. When facing the weeding task of a large area of fields, there are problems such as long operation time and unreasonable resource utilization.

[0003] With the development of agricultural intelligence, the collaborative operation of multiple weeding machines has become an effective way to improve weeding efficiency and quality. However, currently in a large field environment, there are still many deficiencies in the collaborative scheduling control of multiple weeding machines: it is difficult to reasonably allocate the priority operation areas of weeding machines according to complex situations such as the terrain of the large field and the distribution of weeds; and each weeding machine cannot perform efficient weeding operations according to the actual power consumption situation, and cannot dynamically adjust the weeding operation speed, resulting in poor collaborative operation effect and low work efficiency; there is no set weeding operation measurement index for work monitoring, which is prone to cause waste and out-of-control of resources.

[0004] Therefore, a collaborative scheduling control method for a weeding machine in a large field environment is needed. Summary of the Invention

[0005] A collaborative scheduling control method for a weeding machine in a large field environment provided by the present invention aims to reasonably allocate the priority operation areas of weeding machines according to complex situations such as the terrain of the large field and the distribution of weeds, achieve precise collaborative operation, and improve the coverage range and efficiency of weeding operations; combine the actual situation and power consumption situation to enable the weeding machine to dynamically adjust the weeding operation speed, so as to improve operation efficiency, reduce energy consumption, and optimize the collaborative operation effect; establish a weeding operation measurement index to monitor and evaluate the quality and efficiency of weeding operations, and reduce the waste of time and resources.

[0006] The technical solution of the present invention is specifically as follows:

[0007] A collaborative scheduling control method for a weeding machine in a large field environment includes the following steps:

[0008] S1. Use a drone equipped with a camera, a multispectral sensor, and different sensor devices to collect information related to the large field and weeding, preprocess the data, establish a neural network model for evaluating the priority of weeding operations, and output the priority evaluation result of weeding operations;

[0009] S2. Obtain the confidence level of the field weeding operation evaluation information based on the output weeding operation priority evaluation information. Meanwhile, set the credibility threshold of the weeding operation evaluation information, and calculate the real-time weeding operation speed by combining the confidence level of the field weeding operation evaluation information with the power consumption.

[0010] S3. Calculate the determination index of the field weeding efficiency by obtaining the field weeding operation speed and the performance of the weeding equipment. If the obtained determination index of the field weeding efficiency exceeds the range of the weeding benchmark efficiency threshold, start the preparatory work countdown.

[0011] Furthermore, step S1 specifically includes:

[0012] The weeding operation priority evaluation neural network model includes an input layer, a convolutional layer, a dynamic analysis layer, a fully connected layer, a comprehensive evaluation layer, and an output layer. Set the input of the weeding operation priority evaluation neural network model as , where represents the preprocessed field-related image information, and represents the preprocessed other field-related data information; the convolutional layer performs a convolution operation on the preprocessed field-related image information; transfer the extracted frame features of the convolutional layer to the dynamic analysis layer; in the comprehensive evaluation layer, comprehensively evaluate the priority of the field weeding block by combining the results of the dynamic analysis layer and the preprocessed other field-related data information; then enter the fully connected layer, and the fully connected layer outputs the evaluation result of the field situation, and finally the output layer outputs the overall weeding priority evaluation result.

[0013] Furthermore, in the dynamic analysis layer, the specific process is as follows:

[0014] where represents the output of the dynamic analysis layer, indicating the analysis result of the field image in the time period; represents the feature vector of the field-related image; represents the state of the field image in the time period before the start of the record; represents the th feature sensitivity; represents the th feature standard reference value; represents the weight value at different times affected by the climate; represents the projection conversion parameter; represents the element value at the vegetation index index location in the th area at represents the gradient amplitude at the vegetation index positioning segmentation in the th area in the Indicates the error coefficient; Indicates the dynamic positioning parameter.

[0015] Furthermore, step S2 specifically includes:

[0016] Define the first confidence threshold comparison parameter for the weeding operation evaluation information , the second confidence threshold comparison parameter , where , compare the confidence level of the weeding operation evaluation information with the first confidence threshold comparison parameter and the second confidence threshold comparison parameter respectively.

[0017] Furthermore, when the confidence level of the obtained weeding operation evaluation information , the weeding operation speed correction parameter value obtained is , ;

[0018] When the confidence level of the obtained weeding operation evaluation information , the weeding operation speed correction parameter value obtained is , ；

[0019] When the confidence level of the obtained weeding operation evaluation information , the weeding operation speed correction parameter value obtained is ， ;

[0020] where represents the standard weeding operation speed; , respectively represent the influence coefficients of the confidence level on the weeding operation speed within the corresponding threshold intervals.

[0021] Furthermore, step S2 specifically includes:

[0022] , represents a function, representing the mapping relationship between the confidence level of the weeding operation evaluation information and the weeding operation speed ; define the power consumption as , calculate the real-time weeding operation speed , represents the power consumption regulation coefficient, and the specific process is as follows:

[0023] where represents the first boundary; represents the second boundary.

[0024] Furthermore, step S3 specifically includes:

[0025] Calculate the determination index of the weeding efficiency in the field , and a threshold range of the benchmark weeding efficiency is set. If the obtained determination index of the weeding efficiency in the field exceeds the threshold range of the benchmark weeding efficiency, it enters the work monitoring module, and at the same time, the preparatory work countdown is started.

[0026] Furthermore, the calculation process of the determination index of the weeding efficiency in the field is as follows:

[0027]

[0028] Among them, represents the weight coefficient of the weeding operation speed in the field; represents the weight coefficient of the performance of the weeding equipment; represents the parameter of the weeding operation speed in the field; represents the parameter of the performance of the weeding equipment.

[0029] Furthermore, step S3 specifically includes: defining the comparison parameter of the determination index of the weeding efficiency in the field ;

[0030] When ≥ , the obtained determination index of the weeding efficiency in the field exceeds the threshold range of the benchmark weeding efficiency, and the preparatory work countdown is started; when , the preparatory work countdown is not started.

[0031] Beneficial effects: 1. By establishing a neural network model for evaluating the priority of weeding operations, the control system can evaluate the priority of weeding operations according to the collected data, which helps to determine which areas need to be weeded first, improves the operation efficiency and effect, realizes intelligent scheduling, and maximizes the weeding efficiency; precise monitoring is carried out according to the evaluation results of the operation priority, effectively reducing the repetitive labor and resource waste in agricultural operations, helping the field managers to reduce costs, improve production efficiency, and providing data support for the subsequent control process.

[0032] 2. By considering the confidence level of weeding operation evaluation information and the power consumption situation, the present invention adjusts the weeding operation speed according to the situation, thereby improving the stability of the control system; dynamically adjusting the weeding operation speed can optimize the system performance, ensure more accurate weeding operation evaluation results under different power consumption and evaluation situations, effectively save the use of control system resources, and avoid high-speed weeding operations and processing under unnecessary circumstances. At the same time, by dynamically controlling the working state of the device and only entering the preparation and working state when the task of obtaining weeding machine information is required, energy can be effectively saved; when there is no task, the device can enter the low-power initial start state to reduce unnecessary energy consumption; it can also improve the response speed of the system and ensure that the weeding machine information can be obtained in a timely manner when needed.

[0033] 3. The present invention dynamically controls the weeding control work in the field according to the weeding efficiency measurement indicators in the field, discovers problems in a timely manner and takes measures to ensure that the weeding operation is always in an efficient state; by real-time monitoring and adjusting the weeding operation speed and equipment performance, the efficiency of weeding in the field can be maximized, and time and resource waste can be reduced; at the same time, scheduling and controlling the weeding operation according to the actual situation can make more effective use of weeding equipment and human resources, thereby reducing costs and increasing output; by using an automated scheduling control method to reduce human errors and waste, ensuring that the weeding operation in the field is carried out in the best way, and improving the operation quality and efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a flowchart of a collaborative scheduling control method for a weeding machine in a field environment according to the present invention;

[0035] Figure 2 is a module diagram of a collaborative scheduling control system for a weeding machine in a field environment according to the present invention;

[0036] Figure 3 is a schematic diagram of an intelligent weeding control processing unit according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] In order to better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific embodiments. At the same time, it should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0038] Referring to the attached Figure 1 , this embodiment provides a collaborative scheduling control method for a weeding machine in a field environment, including the following steps:

[0039] S1. Use a drone equipped with a high-definition camera, a multispectral sensor, and various related sensors and other monitoring devices to collect field and weed control-related information and preprocess the data, establish a neural network model for evaluating the priority of weed control operations, and output the priority evaluation results of weed control operations.

[0040] Use a drone equipped with a high-definition camera and a multispectral sensor to conduct an all-round scan of the field to obtain topographical information, crop distribution information, and weed distribution information of the field; among them, the high-definition camera is used to take visible light images of the field, and the distribution positions of crops and weeds are identified through image analysis; the multispectral sensor is used to collect spectral data in different bands, and according to the differences in the reflection characteristics of crops and weeds in different spectral bands, more accurately determine the distribution range and density of weeds. At the same time, use various related sensors to obtain meteorological data, soil data, weed control machine working data, vegetation index data, crop growth data, etc., to provide data support for subsequent modeling and analysis, thereby improving the production efficiency of weed control.

[0041] At the same time, transmit the collected image and spectral data to the ground control center. The ground control center uses geographic information system (GIS) technology to process and analyze the data to generate a three-dimensional terrain model, a crop distribution map, and a weed density distribution map of the field. Then, preprocess the obtained field-related information using existing technologies, establish a neural network model for evaluating the priority of weed control operations, input the preprocessed field-related images and other relevant information such as the field soil and vegetation into the neural network model for evaluating the priority of weed control operations, and finally output the corresponding priority evaluation results of weed control operations through deep learning.

[0042] The neural network model for evaluating the priority of weed control operations includes an input layer, a convolutional layer, a dynamic analysis layer, a fully connected layer, a comprehensive evaluation layer, and an output layer. Set the input of the neural network model for evaluating the priority of weed control operations as Among them, represents the preprocessed field-related image information, represents the preprocessed other relevant data information of the field, The convolutional layer performs a convolutional operation on the preprocessed field-related image information to obtain the feature vector of the field-related image. The specific process is as follows:

[0043] Among them, represents the feature vector of the field-related image; represents the convolutional function; represents the weight value; represents the bias of the convolutional layer; transfer the image features extracted by the convolutional layer to the dynamic analysis layer, and further dynamically analyze the field situation in the dynamic analysis layer. The specific process is as follows:

[0044] Among them, represents the output of the dynamic analysis layer, indicating the analysis result of the large field image during the time period; represents the state of the large field image in the time period before the start of recording; represents the sensitivity of the th feature; represents the standard reference value of the th feature; represents the weight value at different times affected by the climate; represents the projection conversion parameter; represents the element value at the vegetation index index location in the th area at ; represents the gradient amplitude at the vegetation index location segmentation in the th area during the time period; represents the error coefficient;

[0045] In the comprehensive evaluation layer, combining the results of the dynamic analysis layer and other relevant data information of the preprocessed field to comprehensively evaluate the priority of the large field weeding block, the specific process is as follows:

[0046] Among them, represents the output result of the comprehensive evaluation layer; represents the soil nutrient content; represents the proportionality coefficient, indicating the influence of soil nutrient changes on weed growth state changes; represents the vegetation or crop coverage rate; represents the time correction factor, used to adjust the changes in other relevant data information of the field over time; represents the sensitivity coefficient; represents the amplification coefficient, which amplifies the influence of the analysis result of the large field image on the comprehensive evaluation process; represents the weighting coefficient of sunlight on the comprehensive evaluation process; represents the sunlight intensity.

[0047] Then it enters the fully connected layer and the Softmax function. The fully connected layer outputs the evaluation result of the large field situation, and finally the output layer outputs the overall weeding priority evaluation result, and the output result is . In the embodiments of the present invention, for the large field blocks with priority weeding operations evaluated , its corresponding output is . After passing through the Softmax function, the confidence of the large field blocks for weeding operations can be expressed as:

[0048] Wherein, represents the number of large field blocks that need to perform weeding operations; represents the traversal variable.

[0049] Then, the error is calculated through the loss function, and the parameters in the neural network for evaluating the priority of weeding operations are optimized through gradient descent, and a neural network model for accurately evaluating the priority of weeding operations is trained.

[0050] By establishing a neural network model for evaluating the priority of weeding operations in the present invention, the control system can evaluate the priority of weeding operations according to the collected data, which helps to determine which areas need to be weeded first, improves the operation efficiency and effect, realizes intelligent scheduling, and maximizes the weeding efficiency; precise monitoring is carried out according to the evaluation results of the operation priority, effectively reducing the repetitive labor and resource waste in agricultural operations, helping the large field manager to reduce costs, improve production efficiency, and providing data support for the subsequent control process.

[0051] S2. Obtain the confidence of the large field weeding operation evaluation information according to the output weeding operation priority evaluation information , and at the same time set the credibility threshold of the weeding operation evaluation information , and calculate the weeding operation speed from the license plate recognition information, that is, the confidence of the license plate information , wherein, define the correction parameter value of the weeding operation speed ,

[0052] Define the first credibility threshold comparison parameter and the second credibility threshold comparison parameter of the weeding operation evaluation information, wherein, , compare the confidence of the weeding operation evaluation information with the first credibility threshold comparison parameter and the second credibility threshold comparison parameter respectively, and obtain the correction parameter value of the weeding operation speed according to the comparison result, , represents the standard weeding operation speed, and the specific process is as follows:

[0053] When the obtained confidence of the weeding operation evaluation information, the correction parameter value of the weeding operation speed is obtained as , ;

[0054] When the confidence level of the weed control operation evaluation information obtained , the weed control operation speed correction parameter value obtained is , ；

[0055] When the confidence level of the weed control operation evaluation information obtained , the weed control operation speed correction parameter value obtained is ， ;

[0056] Among them, and respectively represent the influence coefficients of the confidence level on the weed control operation speed within the corresponding threshold intervals.

[0057] Among them, , the value for increasing the weed control operation speed. The system believes that the confidence level of the current weed control operation evaluation information is relatively high, but there is still room for improvement. For example, in the case where the weed coverage in the field is relatively good but there may be some minor interferences (such as environmental impacts), increasing the weed control operation speed can increase the chance of obtaining more accurate weed control efficiency, thereby improving the overall evaluation effect;

[0058] , the value for keeping the weed control operation speed unchanged, indicating that the confidence level of the weed control operation evaluation information is in a middle range. The system believes that the current weed control operation speed is appropriate and does not need to be adjusted. For example, when the weed coverage in the field basically meets the requirements but there is still some room for improvement, keeping the current weed control operation speed can continue to obtain the weed coverage for evaluation while avoiding system fluctuations caused by unnecessary weed control speed adjustments;

[0059] , the value for decreasing the weed control operation speed, meaning that the system believes that the confidence level of the current weed control operation evaluation information is relatively low. It may be due to abnormal vegetation index data or unstable meteorological information, resulting in a decrease in the confidence level. The system may reduce the speed of the weed control machine to reduce possible incorrect operations and ensure the accuracy of the operation.

[0060] At the same time, the original relationship between the confidence level of the weed control operation evaluation information and the weed control operation speed is: , represents a function, indicating the mapping relationship between the confidence level of the weed control operation evaluation information and the weed control operation speed ; Define the power consumption as , calculate the real-time weeding operation speed , represents the electric energy consumption regulation coefficient, and the specific process is as follows:

[0061] Among them, represents the first boundary; represents the second boundary; is within the normal range of electric energy consumption; represents a large electric energy consumption; represents a small electric energy consumption.

[0062] In the embodiment of the present invention, the measuring device records the weeding operation speed in the field and enters the preparatory work countdown. If the monitoring device detects a task of obtaining weeding machine information, it is in the preparation state, and the preparatory work countdown starts to count down. If the preparatory work countdown reaches zero, the work monitoring module enters the working state. If the monitoring device detects that there is no task of obtaining weeding machine information, the preparatory work countdown enters the initial start state. In the embodiment of the present invention, for example, the preparatory work countdown is set to 8 seconds. When a task of obtaining weeding machine information is detected, the preparatory work countdown starts to count down, such as: 8,..., 5,..., 3,..., 1..., when the value reaches zero, the work monitoring module enters the working state.

[0063] By considering the confidence level of the weeding operation evaluation information and the electric energy consumption situation, the present invention can adjust the weeding operation speed according to the situation, thereby improving the stability of the control system; dynamically adjusting the weeding operation speed can optimize the system performance, ensure more accurate weeding operation evaluation results under different electric energy consumption and evaluation situations, effectively save the use of control system resources, and avoid high-speed weeding operations and processing under unnecessary circumstances. At the same time, by dynamically controlling the working state of the device and only entering the preparation and working states when a task of obtaining weeding machine information is required, energy can be effectively saved; when there is no task, the device can enter the low-power initial start state to reduce unnecessary energy consumption; it can also improve the response speed of the system and ensure that the weeding machine information can be obtained in a timely manner when needed.

[0064] S3. Obtain the weeding operation speed in the field and the performance of the weeding equipment , calculate the determination index of the weeding efficiency in the field , if the determination index of the weeding efficiency in the field exceeds the weeding benchmark efficiency threshold range, start the preparatory work countdown.

[0065] Obtain the weeding operation speed in the field in the weeding information collection unit in the field and the performance of the weeding equipment , in the embodiments of the present invention, the field weeding operation speed represents the speed of completing the removal of weeds in the field within a unit time; the performance of the weeding equipment represents the working efficiency and accuracy of the equipment or machine for the weeding operation, and is measured by the maintenance frequency of the weeding equipment.

[0066] Calculate the determination index of the field weeding efficiency according to the detected data obtained. , and there is a threshold range of the weeding reference efficiency inside the collaborative control module. If the obtained determination index of the field weeding efficiency exceeds the threshold range of the weeding reference efficiency, it enters the working monitoring module, and at the same time, starts the preparatory work countdown. In the embodiments of the present invention, the calculation process of the determination index of the field weeding efficiency is as follows:

[0067] Wherein, represents the weight coefficient of the field weeding operation speed; represents the weight coefficient of the performance of the weeding equipment; represents the parameter of the field weeding operation speed; represents the parameter of the performance of the weeding equipment.

[0068] The specific judgment process is that there is a comparison parameter of the determination index of the field weeding efficiency inside the collaborative control module ;

[0069] When ≥ , the obtained determination index of the field weeding efficiency exceeds the threshold range of the weeding reference efficiency, starts the preparatory work countdown, and enters the working state;

[0070] When , the preparatory work countdown is not started. At the same time, the weeding efficiency compensation module starts to improve the operation speed and equipment performance information through the existing technologies known to those skilled in the art, such as: optimizing the operation path planning, adapting the maintenance frequency, etc.

[0071] The present invention dynamically controls the field weeding control work according to the determination index of the field weeding efficiency, discovers problems in time and takes measures to ensure that the weeding operation is always in an efficient state; by real-time monitoring and adjusting the weeding operation speed and equipment performance, the efficiency of the field weeding can be maximized, and the waste of time and resources can be reduced; at the same time, scheduling and controlling the weeding operation according to the actual situation can make more effective use of the weeding equipment and human resources, thereby reducing costs and increasing output; by the automated scheduling control method, human errors and wastes are reduced, ensuring that the field weeding operation is carried out in the best way, and improving the operation quality and efficiency.

[0072] Refer to the appendix Figure 2, this embodiment provides a collaborative scheduling control system for a weeding machine in a field environment, including the following:

[0073] A field weeding information collection unit, a weeding operation priority evaluation unit, and an intelligent weeding control processing unit;

[0074] The field weeding information collection unit is used to collect various information in the field in real time through various sensors, monitoring devices, drones and other technologies, such as crop growth conditions, weed distribution, soil properties, etc.; obtain data and transmit it to the system to provide support for subsequent analysis and decision-making.

[0075] The weeding operation priority evaluation unit is used to establish a weeding operation priority evaluation neural network model through technologies such as preprocessing and data analysis to evaluate the weeding requirements in different areas of the field;

[0076] Refer to the appendix Figure 3 , the intelligent weeding control processing unit is used to realize the intelligent scheduling and control of the weeding robot according to the priority evaluation results of the weeding operation;

[0077] Among them, the intelligent weeding control processing unit includes a weeding operation speed acquisition module, an electric energy consumption module, a work monitoring module, a weeding efficiency compensation module, and a collaborative control module;

[0078] The weeding operation speed acquisition module is responsible for real-time monitoring and obtaining the weeding machine operation speed data, adjusting the speed according to the operation conditions in different areas of the field to ensure the efficient progress of the weeding operation;

[0079] The electric energy consumption module is used to monitor and evaluate the electric energy consumption of the weeding robot and perform electric energy management and optimization;

[0080] The work monitoring module is used to monitor the working state and efficiency of the weeding machine, enter the working state in time or discover problems and take corresponding measures to ensure that the weeding operation is carried out according to the plan;

[0081] The weeding efficiency compensation module is used to perform compensation adjustment according to the actual weeding efficiency situation, and make corresponding adjustments according to the feedback of the field weeding efficiency measurement index to improve the weeding efficiency and ensure the operation quality;

[0082] The collaborative control module is used to work collaboratively with other system modules to realize the intelligent control and scheduling of the overall weeding operation.

[0083] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It should be understood that each flow and / or block in the flowcharts and / or block diagrams, and combinations of flows and / or blocks in the flowcharts and / or block diagrams can be implemented by computer program instructions. These computer program instructions can be provided to the processors of general purpose computers, special purpose computers, embedded processors, or other programmable data processing devices to produce a machine, such that the instructions executed by the processors of the computer or other programmable data processing devices generate means for implementing the functions specified in one flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0084] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufacture including instruction means that implement the functions specified in one flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0085] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one flow Figure 1 one flow or more flows and / or blocks Figure 1 one block or more blocks.

[0086] Although the preferred embodiments of the present invention have been described, those skilled in the art can make additional changes and modifications once they learn the basic creative concepts. Therefore, the appended claims are intended to be construed to include the preferred embodiments as well as all changes and modifications falling within the scope of the present invention.

[0087] The above content is only to illustrate the technical idea of the present invention and cannot limit the protection scope of the present invention. Any modification made on the basis of the technical solution according to the technical idea proposed by the present invention falls within the protection scope of the claims of the present invention.

Claims

1. A coordinated dispatching control method for a weeder in a field environment, characterized in that: The following steps are involved: S1. Use drones equipped with cameras, multispectral sensors and different sensor devices to collect field and weeding related information and pre-process the data, establish a neural network model for weeding operation priority evaluation, and output the weeding operation priority evaluation results; The neural network model for weeding operation priority evaluation includes input layer, convolution layer, dynamic analysis layer, fully connected layer, comprehensive evaluation layer, and output layer. The input of the neural network model for weeding operation priority evaluation is ,in, Represents the preprocessed field-related image information, Represents other relevant data information of the field after preprocessing; the output layer outputs the weeding priority evaluation results as a whole, and the output result is , for the field blocks that have been evaluated as priority for weed control , and its corresponding output is ; In the dynamic analysis layer, the specific process is as follows: in, Represents the output of the dynamic analysis layer, which is represented in The analysis results of field images during different time periods; A feature vector representing a field-related image; Indicates the state of the field image in the period before the recording starts; Indicates The sensitivity of a feature; Indicates Standard reference value for each characteristic; Indicates the weight value at different times under the influence of climate; Represents projection transformation parameters; express At the The element value at the location of the regional vegetation index index; express The period is The gradient amplitude at the location segmentation of the regional vegetation index; represents the error coefficient; Indicates dynamic positioning parameters; S2. Obtain the confidence of the field weeding operation assessment information according to the output weeding operation priority assessment information, set the credibility threshold of the weeding operation assessment information at the same time, and calculate the real-time weeding operation speed by combining the confidence of the field weeding operation assessment information with the power consumption; S3. Calculate the field weeding efficiency measurement index by obtaining the field weeding operation speed and the performance of the weeding equipment. If the field weeding efficiency measurement index exceeds the weeding benchmark efficiency threshold range, start the countdown for the preparatory work.

2. The coordinated dispatching control method of a weeder in a field environment according to claim 1 is characterized in that: The step S1 specifically includes: The convolution layer performs convolution operations on the preprocessed field-related image information; the image features extracted by the convolution layer are passed to the dynamic analysis layer; in the comprehensive evaluation layer, the priority of the field weeding blocks is comprehensively evaluated by combining the results of the dynamic analysis layer and other relevant data information of the preprocessed field; then it enters the fully connected layer, which outputs the evaluation results of the field conditions, and finally the output layer outputs the weeding priority evaluation results as a whole.

3. The coordinated dispatching control method of a weeder in a field environment according to claim 1, characterized in that: The step S2 specifically includes: Define the first credibility threshold comparison parameter of weeding operation evaluation information , the second credibility threshold comparison parameter ,in, , the confidence level of the weeding operation assessment information Compare the parameters with the first credibility threshold respectively Compare the parameter with the second credibility threshold for comparison.

4. The coordinated dispatching control method of a weeder in a field environment according to claim 3 is characterized in that: The step S2 specifically includes: obtaining a weeding operation speed correction parameter value according to the comparison result: , , Indicates the standard weeding operation speed; When the confidence level of the weed control operation assessment information is , then the weeding operation speed correction parameter is , ; When the confidence level of the weed control operation assessment information is , then the weeding operation speed correction parameter is , ； When the confidence level of weed control assessment information is obtained , then the weeding operation speed correction parameter is ， ; in, , They respectively represent the influence coefficient of built-in confidence on the weeding operation speed in the corresponding threshold interval.

5. The coordinated dispatching control method of a weeder in a field environment according to claim 4 is characterized in that: The step S2 specifically includes: , Represents a function that represents the confidence of the weeding operation evaluation information and weeding speed The mapping relationship between them; define the power consumption as , calculate the real-time weeding operation speed , Represents the power consumption control coefficient. The specific process is as follows: in, represents the first boundary; Indicates the second boundary.

6. The coordinated dispatching control method of a weeder in a field environment according to claim 1, characterized in that: The step S3 specifically includes: Calculation of field weed control efficiency test index , and a threshold range of weed control efficiency is set. If the field weed control efficiency determination index is If the weeding benchmark efficiency threshold is exceeded, the work monitoring module is entered and the preparatory work countdown is started.

7. The coordinated dispatching control method of a weeder in a field environment according to claim 6, characterized in that: The calculation process of the field weed control efficiency determination index is as follows: in, The weight coefficient representing the speed of field weeding operation; A weight coefficient indicating the performance of the weeding equipment; Indicates the speed parameter of field weeding operation; Indicates the performance parameters of weeding equipment.

8. The coordinated dispatching control method of a weeder in a field environment according to claim 6, characterized in that: The step S3 specifically includes: defining the comparison parameters of the field weed control efficiency determination index ; when ≥ When the field weed control efficiency is determined, the field weed control efficiency index is obtained. When the threshold of the weed control efficiency is exceeded, the countdown for the preparation work is started; The countdown for preparation will not start.

Images